Thanks for the suggestion which I followed. The size of Weather is the same on both ends, and the values received are just as expected. There is nothing wrong with the data structure. Here's the problem ... when my laptop is outside connected to the nano to trouble shoot, all is well. When I leave and plug the nano into a 5v transformer to the USB input - the serial leads are shorted or have too much noise (or floating ground issue?) - and the serial data is not transmitted. Serial.available() test is never true and thus the pool data is not read or loaded into the structure. Since there is always a fresh power up when I leave - the values are zero when I get inside. I was tipped to the problem by the .pump flag which once set inside does not reset. When I canned in values to debug it set the flag true so I knew the transmission had worked at least once. Solution will be to power the device by other than by the USB port. If I could have the hours back I've spent on this issue .... ? Oh well.